[03:26:09] <wikibugs>	 10ORES, 10Scoring-platform-team (Current), 10Operations: ORES uwsgi consumes a large amount of memory and CPU when shutting down (as part of a restart) - https://phabricator.wikimedia.org/T242705 (10Gilles) @dpifke has looked into this. Dave, can you share notes here about what you've tried and what you unde...
[09:20:47] <wikibugs>	 10ORES, 10Scoring-platform-team, 10Operations: Move ORES to redis misc cluster - https://phabricator.wikimedia.org/T254226 (10akosiaris)
[09:20:56] <wikibugs>	 10ORES, 10Scoring-platform-team, 10Operations: Move ORES to redis misc cluster - https://phabricator.wikimedia.org/T254226 (10akosiaris) p:05Triage→03High
[09:25:03] <wikibugs>	 10ORES, 10Scoring-platform-team, 10Operations, 10Patch-For-Review: Move ORES to redis misc cluster - https://phabricator.wikimedia.org/T254226 (10akosiaris)
[10:26:25] <wikibugs>	 10ORES, 10Scoring-platform-team, 10Operations: Move ORES to redis misc cluster - https://phabricator.wikimedia.org/T254226 (10akosiaris) codfw migration has gone really well, I 've barely managed to notice the migration in the dashboards.
[10:56:05] <wikibugs>	 10ORES, 10Scoring-platform-team, 10Operations: Move ORES to redis misc cluster - https://phabricator.wikimedia.org/T254226 (10akosiaris) eqiad migrations has gone pretty much as well. There seem to be some occasional overloads due to ores1001 at some point, it looks like a restart of uwsgi+celery fixed it.
[11:17:34] <wikibugs>	 10ORES, 10Scoring-platform-team, 10Operations: Move ORES to redis misc cluster - https://phabricator.wikimedia.org/T254226 (10akosiaris) 05Open→03Stalled https://grafana.wikimedia.org/d/RLhtAw6mz/ores-redis?orgId=1&refresh=1m has been updated as well. I am gonna call this resolved, but we should wait a c...
[12:57:14] <wikibugs>	 10ORES, 10Scoring-platform-team, 10Operations: Move ORES to redis misc cluster - https://phabricator.wikimedia.org/T254226 (10akosiaris)
[15:42:31] <halfak>	 https://commons.wikimedia.org/wiki/File:ORES_-_Lowering_Barriers_with_Participatory_Machine_Learning_in_Wikipedia.pdf
[15:42:37] <halfak>	 It is submitted!
[15:43:53] <halfak>	 chtnnh, ^ 
[15:44:33] <chtnnh>	 Yay!! Kudos to you halfak for working against all odds on this :D
[15:44:58] <chtnnh>	 brb gotta reboot
[15:45:11] <halfak>	 39 pages of descriptions of machine learning, meaning negotiation, technological strategies, and implications for design. 
[16:15:22] <wikibugs>	 10Scoring-platform-team, 10Research: Write Python util for converting Wikidata claims to features for ML models - https://phabricator.wikimedia.org/T252775 (10Isaac)
[18:44:28] <haksoat>	 halfak: I'm having this weird issue with flake on articlequality
[18:44:46] * halfak looks
[18:44:48] <haksoat>	 I get:
[18:44:51] <haksoat>	 W504 line break after binary operator
[18:44:58] <haksoat>	 But then when I do the break before
[18:45:05] <halfak>	 Ahh yes.  It looks like a new version of flake8 has added some new warnings.  How annoying. 
[18:45:11] <haksoat>	 I get
[18:45:13] <haksoat>	 W503 line break before binary operator
[18:45:29] <haksoat>	 Yeah
[18:45:32] <halfak>	 Yeah.  That warning is nonsense and I wish they would remove it from flake8
[18:45:42] <halfak>	 I'll show you how to get around this. 
[18:45:47] <haksoat>	 Now it fails doing breaks before and after
[18:45:52] <haksoat>	 Okay. Thanks.
[18:46:22] <halfak>	 Add these exceptions: E741, W504, F841
[18:46:28] <halfak>	 To this file: 
[18:46:42] <halfak>	 https://github.com/wikimedia/articlequality/blob/master/tox.ini
[18:46:49] <halfak>	 And that will work for now. 
[18:47:40] <haksoat>	 Okay
[19:00:27] <haksoat>	 Issue persists
[19:00:40] <haksoat>	 I don't think it sees the tox.ini file
[19:00:45] <haksoat>	 halfak: 
[19:01:20] <halfak>	 Hmm.  It definitely *should*
[19:01:23] <halfak>	 Works in other repos. 
[19:02:25] <haksoat>	 I have the exceptions added
[19:02:26] <haksoat>	 https://github.com/wikimedia/articlequality/blob/173836403f35c669e2ca6b782870fc87c147e79b/tox.ini
[19:02:36] <haksoat>	 Error: https://travis-ci.org/github/wikimedia/articlequality/builds/693975978
[19:05:35] * halfak checks it out
[19:06:23] <halfak>	 Aha!  It looks like the --ignore in the flake8 call is overruling. 
[19:06:54] <halfak>	 "flake8 . --exclude=.svn,CVS,.bzr,.hg,.git,__pycache__,.tox,.eggs,*.egg,docs,examples"
[19:07:14] <halfak>	 in .travis.yml
[19:07:26] <halfak>	 Notice that I dropped the --ignore line. 
[19:07:30] <halfak>	 That shouldn't have been there./ 
[19:08:05] <haksoat>	 Oh. Okay.
[19:17:40] <halfak>	 accraze, got a few minutes to talk about planning re. ORES 2.0?
[19:21:37] <haksoat>	 Works
[19:21:46] <haksoat>	 https://github.com/wikimedia/articlequality/pull/134
[19:21:51] <haksoat>	 halfak: tests pass
[19:23:25] <halfak>	 Nice. 
[19:33:54] <halfak>	 haksoat, do you have access to ores-misc-01?
[19:34:01] <halfak>	 I'd like to have you rebuild the model. 
[19:41:34] <haksoat>	 Yes. I do.
[19:41:43] <haksoat>	 Okay. 
[19:42:03] <haksoat>	 The one for article quality right?
[19:42:15] <haksoat>	 halfak: 
[19:52:35] <halfak>	 Right. 
[19:52:55] <halfak>	 haksoat, unrelated, have you had a chance to look at https://gist.github.com/halfak/f06ac6a47c2bcf5dfda545c55eb0a1a5
[19:53:01] <halfak>	 It looks like it breaks our tokenizer. 
[20:00:55] <haksoat>	 Taking a look
[20:03:15] <haksoat>	 Error is in Japanese
[20:03:37] <halfak>	 Error? 
[20:03:46] <halfak>	 When I run it, it just spins forever
[20:03:53] <haksoat>	 Oh
[20:04:03] <haksoat>	 I was looking at the revision page
[20:04:12] <haksoat>	 I'll run here
[20:09:09] <haksoat>	 Same experience
[20:10:20] <halfak>	 I figure it's got to be something of a breakdown in the regex.  Like, some pattern just falls down trying to match recursively or something. 
[20:11:11] <haksoat>	 Yeah. Taking a look.
[20:16:14] <accraze>	 halfak any chance we can talk ORES 2.0 tomorrow or thurs? I don
[20:16:35] <accraze>	 im still taking care of stuff at home right now
[20:17:01] <halfak>	 Dang.  It's time sensitive.  I could probably keep it really quick if you could squeeze in 5 minutes.  Sorry for the trouble :( 
[20:18:10] <accraze>	 sure that's fine
[20:18:33] <halfak>	 OK call when you get a chance.  I'll be around. 
[20:18:47] <accraze>	 cool gimmie just a sec
[20:26:13] <haksoat>	 The ref_singleton part of the regex is the one with the problem
[20:26:24] <wikibugs>	 10Scoring-platform-team (Research), 10Structured-Data-Backlog, 10artificial-intelligence: Implement NSFW image classifier using Open NSFW - https://phabricator.wikimedia.org/T214201 (10Nuria) FYI that there is some talk about WMF deploying a NSFW classifier for commons on the year 2020-2021
[20:51:09] <halfak>	 haksoat, aha.  
[20:51:23] <halfak>	 I need to go AFK for a bit.  Back in about an hour. 
[21:22:20] <haksoat>	 Okay
[21:22:42] <haksoat>	 I've opened a PR for it though halAFK so you can take a look when back
[21:22:44] <haksoat>	 https://github.com/halfak/deltas/pull/13
[21:46:22] <wikibugs>	 10Scoring-platform-team, 10Research: Add wikidata to articletopic pipeline - https://phabricator.wikimedia.org/T254289 (10Isaac)
[21:52:19] <halfak>	 Merged!
[21:52:57] <haksoat>	 Nice
[21:53:34] <haksoat>	 I'll take a look at building the models for articlequality tomorrow
[21:53:39] <haksoat>	 Gotta go now
[21:53:43] <haksoat>	 :)
[22:05:00] <halfak>	 Just released deltas 0.5.1
[22:05:20] <halfak>	 We should be able to pull that into revscoring.  I can use that to retry the model rebuild tomorrow.